Safety and efficacy of mechanical thrombectomy in infective endocarditis: A matched case-control analysis from the German Stroke Registry-Endovascular Treatment.

BACKGROUND AND PURPOSE: Up to 30% of infective endocarditis (IE) patients have ischemic stroke as a complication. Standard treatment with mechanical thrombectomy (MT) with or without intravenous thrombolysis for large vessel occlusion (LVO) has not been evaluated formally in these patients.

METHODS

Patients enrolled in the German Stroke Registry-Endovascular Treatment (GSR-ET) between June 2015 and December 2019 were analyzed. Patients with stroke due to IE and patients with cardioembolic stroke and atrial fibrillation (AF) were compared using propensity score matching. Successful reperfusion was defined as modified Thrombolysis in Cerebral Infarction score = 2b-3. Modified Rankin Scale (mRS) = 0-2 at 3 months indicated good outcome.

RESULTS

Of 6635 patients, 55 patients (age = 69.0 ± 13.3 years, 43.6% female, median premorbid mRS (pmRS) = 1, interquartile range [IQR] = 0-1, National Institutes of Health Stroke Scale [NIHSS] = 15, IQR = 10-21) presented with septic embolic stroke due to IE and were compared to 104 patients (age = 66.5 ± 13.4 years, 39.4% female, pmRS = 0, IQR = 0-2, NIHSS = 16, IQR = 10-20) with cardioembolic stroke due to AF. Successful recanalization was achieved in 74.5% of endocarditis patients compared to 87.5% of controls (p = 0.039). Intracranial hemorrhage rates were comparable (30.9% vs. 21.6%, p = 0.175). Good functional outcome was 20.0% in patients with IE compared to 43.3% in matched patients (p = 0.006), with a significantly higher mortality (60.0% vs. 28.8%, p < 0.001). IE was strongly associated with poor outcome (odds ratio [OR] = 0.32, 95% confidence interval [CI] = 0.11-0.87, p = 0.03 for good outcome) and mortality (OR = 4.49, 95% CI = 1.80-10.68, p = 0.001).

CONCLUSIONS

Although MT results in high successful recanalization rates with acceptable safety profile, patients with LVO stroke due to IE have poor outcome.